The development of quantitative models of binaural information processing is proposed, beginning with the further development and evaluation of a correlation model of binaural analysis (Osman, 1971). The work will aim towards a general theory of binaural detection, recognition, and lateralization. Several models will be evaluated by extensive comparisons with empirical results, and by comparisons with each other. The structure of these models will be analyzed in order to determine what the most reasonable structure of a general model should be with regard to: 1) the functional form of the operations presumed to be carried out by the receiver, and 2) the nature of (necessary) assumptions regarding the structure and role of "internal" noise. An attempt to develop theoretical structures related to neurophysiological findings will be made. Several other related projects are outlined. Experimental work is proposed in relation to the basic theoretical work. A similar project on visual pattern recognition is also planned.